Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. Apparatus for detecting the gaze direction of an eye of a user in a vehicle, the apparatus comprising: an illuminator arranged to direct infrared light towards a detection region, the detection region including the eye of the user and a calibration target at a known location in the vehicle; a plurality of cameras having at least partially overlapping fields of view in the detection region and configured to generate an image of the detection region which includes the calibration target and the eye of the user; a controller configured to detect from the image the calibration target in an overlap region of the image corresponding to the at least partially overlapping fields of view of at least two of the plurality of cameras, and to determine the position and/or orientation of the at least two cameras relative to the calibration target, wherein the controller detects the calibration target from at least two of the plurality of cameras simultaneously; a sensor configured to detect a change in the position and/or orientation of at least one of the plurality of cameras with respect to the calibration target, and upon detection of a change in the position and/or orientation, the controller configured to detect again the calibration target at the known location in an overlap region of an image of the detection region and to determine again the position and/or orientation of the at least one of the plurality of cameras relative to the calibration target, wherein the controller detects again the calibration target from at least two of the plurality of cameras simultaneously; the controller further configured to detect, based on an image from at least one of the plurality of cameras, a gaze direction of an eye of the user of the vehicle.
This invention relates to a system for detecting the gaze direction of a vehicle user's eye using multiple cameras and infrared illumination. The system addresses the challenge of accurately tracking gaze direction in a dynamic vehicle environment where camera positions may shift due to vibrations or movements. The apparatus includes an infrared illuminator that directs light toward a detection region containing the user's eye and a calibration target at a fixed, known location in the vehicle. Multiple cameras with overlapping fields of view capture images of this region, including the calibration target and the user's eye. A controller processes these images to detect the calibration target in the overlap regions of at least two cameras, determining their relative positions and orientations. If a sensor detects any change in camera position or orientation, the controller re-detects the calibration target to recalibrate the system. The controller then analyzes images from the cameras to determine the user's gaze direction. This approach ensures accurate gaze tracking by continuously verifying camera alignment with the calibration target, compensating for environmental disturbances.
2. Apparatus according to claim 1 , wherein the controller is configured to detect a plurality of calibration targets in one image.
The invention relates to an apparatus for detecting and processing calibration targets in imaging systems, particularly in applications requiring precise spatial measurements or alignment. The problem addressed is the need for efficient and accurate detection of multiple calibration targets within a single image, which is critical for tasks such as camera calibration, 3D reconstruction, or robotic navigation. The apparatus includes a controller configured to detect a plurality of calibration targets in one image. The controller processes the image to identify and locate multiple calibration targets simultaneously, improving efficiency and reducing the need for multiple image captures or complex post-processing steps. The calibration targets may include patterns such as checkerboards, circles, or other geometric markers used for spatial referencing. The controller's ability to detect multiple targets in a single image enhances the accuracy of calibration processes by providing a broader set of reference points, which can be used to correct lens distortion, determine camera pose, or align multiple imaging devices. The apparatus may also include an imaging device, such as a camera, that captures the image containing the calibration targets. The controller analyzes the image data to identify the targets based on predefined patterns or features, ensuring robust detection even in varying lighting conditions or with partial occlusions. The system may further include a memory for storing calibration data or a communication interface for transmitting the detected target positions to other systems for further processing. This invention improves the speed and reliability of calibration procedures in applications requiring precise spatial measurements.
3. Apparatus according to claim 1 , wherein the controller comprises a pattern recognizer for recognizing a component of a vehicle as the calibration target.
This invention relates to an apparatus for calibrating a sensor system, particularly for use in autonomous vehicles or advanced driver assistance systems (ADAS). The primary problem addressed is the need for accurate and reliable calibration of sensors, such as cameras or LiDAR, to ensure precise environmental perception and safe vehicle operation. Traditional calibration methods often rely on artificial targets or complex setups, which can be impractical in real-world scenarios. The apparatus includes a controller that identifies and uses a component of a vehicle as a calibration target. This eliminates the need for external calibration tools or markers, simplifying the process. The controller incorporates a pattern recognizer to detect and analyze specific vehicle components, such as headlights, taillights, or license plates, which serve as reference points for calibration. By leveraging existing vehicle features, the system ensures consistent and repeatable calibration without additional infrastructure. The pattern recognizer may use image processing, machine learning, or other recognition techniques to identify these components accurately. This approach enhances calibration efficiency, reduces costs, and improves sensor performance in dynamic environments. The invention is particularly useful in automated driving systems where precise sensor alignment is critical for safety and functionality.
4. Apparatus according to claim 1 , further comprising a calibration target, the calibration target comprising a retro-reflector.
A system for optical measurement or alignment includes a calibration target designed to enhance precision. The calibration target features a retro-reflector, which is a reflective surface that returns incident light back toward its source with minimal scattering. This property allows for accurate positioning and calibration of optical instruments, such as cameras, sensors, or laser systems, by providing a stable reference point. The retro-reflector can be used to verify alignment, correct distortions, or compensate for environmental factors like temperature or vibration. The system may also include a light source and a detector to measure reflected signals, ensuring consistent performance in applications like metrology, robotics, or industrial automation. The retro-reflector's design ensures high reflectivity and minimal angular deviation, improving measurement accuracy and reliability. This calibration target is particularly useful in environments where precise optical alignment is critical, such as in manufacturing, scientific research, or autonomous navigation.
5. A vehicle having installed therein apparatus according to claim 1 .
This invention relates to a vehicle equipped with an apparatus designed to enhance safety and operational efficiency. The apparatus includes a sensor system configured to detect environmental conditions around the vehicle, such as obstacles, weather, or road surface conditions. The sensor system may include cameras, radar, lidar, or other detection technologies to gather real-time data. The apparatus also features a processing unit that analyzes the sensor data to identify potential hazards or operational risks. Based on this analysis, the processing unit generates control signals to adjust vehicle systems, such as braking, steering, or acceleration, to mitigate risks or optimize performance. Additionally, the apparatus may include a user interface to alert the driver or provide recommendations for corrective actions. The vehicle may be an autonomous or semi-autonomous vehicle, where the apparatus operates independently or in conjunction with a human driver. The system aims to improve safety by reducing the likelihood of accidents and enhancing situational awareness for the vehicle operator. The apparatus may also integrate with existing vehicle control systems to ensure seamless operation and compatibility. The overall goal is to provide a robust, adaptive system that responds to dynamic driving conditions to ensure safer and more efficient vehicle operation.
6. A method for calibrating and using an apparatus for detecting the gaze direction of an eye of a user in a vehicle, the method comprising: directing infrared light towards a detection region which includes the eye of the user and a calibration target at a known location in the vehicle; generating an image of the detection region using a plurality of cameras, the plurality of cameras having at least partially overlapping fields of view in the detection region; detecting from the image the calibration target in an overlap region corresponding to the at least partially overlapping fields of view of at least two of the plurality of cameras simultaneously; determining the position and/or orientation of the at least two cameras relative to the calibration target; sensing a change in the position and/or orientation of at least one of the plurality of cameras with respect to the calibration target, and upon sensing of a change in the position and/or orientation, detecting again the calibration target simultaneously at the known location in an overlap region of an image of the detection region and determining again the position and/or orientation of the at least one of the plurality of cameras relative to the calibration target; and detecting, based on an image from at least one of the plurality of cameras, a gaze direction of the user of the vehicle.
This invention relates to a system for calibrating and using an apparatus to detect the gaze direction of a user's eye in a vehicle. The system addresses the challenge of accurately tracking eye gaze in dynamic environments, such as vehicles, where camera positions may shift due to vibrations or movements. The method involves directing infrared light toward a detection region that includes the user's eye and a calibration target placed at a known location in the vehicle. Multiple cameras capture images of this region, with their fields of view at least partially overlapping. The system detects the calibration target in the overlapping regions of at least two cameras simultaneously, allowing it to determine the relative position and orientation of these cameras with respect to the target. If a change in camera position or orientation is sensed, the system recalibrates by detecting the target again and updating the camera positions. Finally, the system uses images from the cameras to detect the user's gaze direction. This approach ensures continuous calibration, improving gaze tracking accuracy in vehicles despite potential camera misalignments.
7. Apparatus according to claim 1 , wherein each of the plurality of cameras are provided in a respective sensor module, each respective sensor module including an image sensor and an infrared light source.
This invention relates to a surveillance or monitoring apparatus designed to enhance visibility in low-light or dark environments. The apparatus addresses the challenge of capturing clear images in poor lighting conditions by integrating infrared (IR) illumination with image capture. The system includes multiple cameras, each housed within a dedicated sensor module. Each sensor module contains an image sensor for capturing visual data and an infrared light source to illuminate the scene. The IR light source emits infrared radiation that reflects off objects in the scene, allowing the image sensor to detect and capture detailed images even in darkness. By incorporating both the image sensor and IR light source within the same module, the apparatus ensures precise alignment between illumination and image capture, improving image quality and reducing glare or distortion. The modular design allows for flexible deployment, enabling the apparatus to be adapted for various surveillance or monitoring applications, such as security systems, wildlife observation, or industrial inspections. The use of infrared technology ensures that the apparatus can operate effectively without relying on external light sources, making it suitable for environments where traditional lighting is impractical or unavailable.
8. Apparatus according to claim 1 , wherein the controller is configured to detect two calibration targets in one image, the two calibration targets being in two orthogonal directions with respect to a predetermined position in the image.
This invention relates to an apparatus for detecting and calibrating calibration targets in an image. The apparatus includes a controller that identifies two calibration targets within a single image, where the targets are positioned in two orthogonal directions relative to a predetermined reference position in the image. The calibration targets are used to establish a reference frame or coordinate system for spatial measurements or alignment in imaging systems. The controller processes the image to locate these targets, ensuring accurate positioning and orientation for applications such as machine vision, robotics, or automated inspection. The orthogonal arrangement of the targets allows for precise determination of both horizontal and vertical alignment, improving calibration accuracy. The apparatus may be part of a larger imaging system, such as a camera or sensor array, where calibration is essential for maintaining measurement consistency. The invention addresses the challenge of efficiently calibrating multiple axes in a single image capture, reducing the need for multiple images or complex adjustments. The controller's ability to detect and analyze the targets in one image streamlines the calibration process, enhancing efficiency and reliability in automated systems.
9. Apparatus according to claim 2 , wherein the controller is configured to determine the position and/or orientation of the at least two cameras based on detected positions of the plurality of calibration targets.
This invention relates to a calibration system for determining the position and orientation of multiple cameras in a three-dimensional space. The system addresses the challenge of accurately aligning and calibrating cameras to ensure precise spatial measurements, which is critical in applications such as augmented reality, robotics, and computer vision. The apparatus includes at least two cameras and a controller. The cameras capture images of a plurality of calibration targets positioned within their field of view. The controller processes these images to detect the positions of the calibration targets and uses this data to determine the position and orientation of each camera relative to the targets. This calibration process ensures that the cameras are properly aligned and can accurately reconstruct the 3D environment. The controller may employ algorithms such as triangulation or perspective-n-point (PnP) to compute the camera positions and orientations based on the detected target positions. The system may also include additional features, such as dynamic adjustment of camera parameters to compensate for environmental changes or movement. By accurately determining the spatial relationship between the cameras and the calibration targets, the system enables high-precision tracking and mapping in various applications.
10. A method according to claim 6 , further comprising detecting a plurality of calibration targets in one image.
A method for improving calibration in imaging systems involves detecting multiple calibration targets within a single image to enhance accuracy and efficiency. The method addresses the challenge of precise calibration in imaging applications, such as robotics, autonomous navigation, or industrial inspection, where misalignment or distortion can degrade performance. By identifying and analyzing multiple calibration targets in one image, the system can refine spatial relationships, correct distortions, and optimize alignment more effectively than traditional single-target approaches. The method leverages image processing techniques to locate and analyze the targets, ensuring robust calibration even in dynamic or cluttered environments. This approach reduces the need for multiple images or complex setups, streamlining the calibration process while maintaining high precision. The method is particularly useful in applications requiring real-time adjustments or high-accuracy measurements, such as 3D reconstruction, augmented reality, or automated quality control. By integrating multiple targets into a single image, the system achieves faster, more reliable calibration, improving overall system performance and reducing errors.
11. A method to claim 10 , further comprising determining the position and/or orientation of the at least two cameras based on detected positions of the plurality of calibration targets.
This invention relates to a method for calibrating a multi-camera system, particularly for determining the spatial relationship between multiple cameras. The problem addressed is the need for accurate positioning and orientation of cameras in systems where multiple cameras are used, such as in 3D reconstruction, augmented reality, or robotic vision. Misalignment or incorrect calibration can lead to errors in depth perception, object tracking, or spatial mapping. The method involves using a plurality of calibration targets placed within the field of view of at least two cameras. These targets are detected in the images captured by the cameras, and their positions are analyzed to determine the relative position and orientation of the cameras. By comparing the detected positions of the calibration targets across the different camera views, the system can compute the spatial transformation between the cameras. This allows for precise alignment of the cameras, ensuring accurate depth estimation, object localization, and spatial consistency in the combined output. The calibration targets may include known patterns or markers, such as checkerboards, dots, or other identifiable features, which are easily detectable in the images. The method may also involve solving a geometric optimization problem to refine the camera positions and orientations based on the detected target positions. This ensures robustness against noise and measurement errors, improving the overall accuracy of the calibration. The technique is applicable in various fields, including autonomous vehicles, surveillance systems, and industrial automation, where precise multi-camera calibration is essential.
12. A method according to claim 10 , further comprising detecting tow calibration targets in one image, the two calibration targets being in two orthogonal directions with respect to a predetermined position in the image.
This invention relates to a method for calibrating a camera system, particularly for detecting and processing calibration targets in an image. The method addresses the challenge of accurately determining the spatial relationship between a camera and its environment by using calibration targets to establish reference points. The invention improves upon existing calibration techniques by incorporating the detection of two calibration targets in a single image, where the targets are oriented in two orthogonal directions relative to a predetermined position within the image. This dual-target approach enhances calibration precision by providing additional reference points that account for variations in camera positioning and orientation. The method involves analyzing the image to identify the two targets, determining their positions, and using these positions to refine the calibration parameters. This ensures more accurate alignment and measurement in applications such as computer vision, robotics, and automated inspection systems. The orthogonal arrangement of the targets allows for independent verification of calibration accuracy in multiple axes, reducing errors and improving system reliability. The method is particularly useful in dynamic environments where precise spatial mapping is critical.
13. A method according to claim 6 , further comprising recognizing a component of a vehicle as the calibration target.
This invention relates to vehicle calibration systems, specifically methods for identifying and using vehicle components as calibration targets during calibration processes. The problem addressed is the need for accurate and efficient calibration of vehicle systems, such as sensors or cameras, by leveraging existing vehicle components as reference points rather than relying on external calibration targets. The method involves recognizing a component of a vehicle, such as a headlight, taillight, or other identifiable part, as the calibration target. This eliminates the need for additional external targets, simplifying the calibration process and reducing setup time. The system may use image recognition, sensor data, or other detection techniques to identify the vehicle component and determine its position, orientation, or other relevant parameters. Once identified, the component serves as a reference point for calibrating vehicle systems, ensuring precise alignment and functionality. The method may also include steps for verifying the accuracy of the calibration by comparing the detected component's properties with predefined standards or historical data. This ensures that the calibration is performed correctly and that the vehicle component is suitable for use as a target. The approach is particularly useful in automated or semi-automated calibration systems, where minimizing manual intervention is desirable. By using existing vehicle components as calibration targets, the method improves efficiency, reduces costs, and enhances the reliability of vehicle calibration processes.
14. A method according to claim 6 , wherein the calibration target comprises a retro-reflector.
A method for calibrating optical systems, particularly in imaging or measurement applications, addresses the challenge of accurately determining the position and orientation of a sensor relative to a calibration target. The method involves using a calibration target with a retro-reflector to enhance precision. The retro-reflector, which reflects incident light back toward its source, improves the accuracy of optical measurements by reducing errors caused by environmental factors like ambient light or misalignment. The calibration process includes capturing images or data from the sensor while the retro-reflector is illuminated, then analyzing the reflected light to determine the sensor's position and orientation. This approach is useful in applications such as 3D scanning, robotics, and automated optical inspection, where precise calibration is critical for performance. The retro-reflector's ability to return light directly to the sensor minimizes distortion and improves signal-to-noise ratio, leading to more reliable calibration results. The method may also include additional steps such as adjusting the sensor's parameters or compensating for environmental conditions to further enhance accuracy.
15. A method according to claim 6 , wherein each of the plurality of the plurality of cameras are provided in a respective sensor module, each respective sensor module including an image sensor and an infrared light source.
This invention relates to a surveillance or monitoring system using multiple cameras, each integrated into a dedicated sensor module. The system addresses the challenge of capturing high-quality images in varying lighting conditions, particularly in low-light environments. Each sensor module contains an image sensor for capturing visual data and an infrared (IR) light source to enhance visibility in darkness. The IR light source illuminates the target area, allowing the image sensor to produce clear images even when ambient light is insufficient. The modular design enables flexible deployment, with each camera operating independently or in coordination with others to cover a broader area. The system may be used in security applications, industrial monitoring, or autonomous navigation, where reliable imaging is critical regardless of lighting conditions. The inclusion of IR illumination ensures consistent performance without relying solely on external lighting sources. This approach improves image quality and reliability in low-visibility scenarios, making it suitable for environments where traditional cameras would struggle. The modular structure also allows for easy maintenance and scalability, as individual sensor modules can be added or replaced without disrupting the entire system.
16. Apparatus for detecting the gaze direction of an eye of a user in a vehicle, the apparatus comprising: an illuminator arranged to direct infrared light towards a detection region, the detection region including the eye of the user and calibration targets at known locations in the vehicle; a plurality of cameras responsive to the light in the detection region and configured to generate an image of the detection region, said detection region image including images of calibration targets at the known locations, the plurality of cameras having at least partially overlapping fields of view in an overlap region of the detection region; a pattern recognition module to recognize the calibration targets; a sensor configured to detect a change in the position and/or orientation of at least one of the plurality of cameras; and a control system responsive to the pattern recognition module, to the sensor and to the cameras and operating suitable programs to: (i) determine simultaneously, for each camera, in two orthogonal directions, pixel coordinates of a center of an image of a first calibration target relative to an origin, (ii) determine simultaneously, for each camera, in two orthogonal directions, pixel coordinates of a center of an image of a second calibration target relative to the origin, and (iii) utilize, for each camera, the pixel coordinates of the first calibration target and of the second calibration target to determine the position and/or orientation of the at least one of the plurality of cameras relative to the calibration targets, (iv) upon the sensor detecting a change in the position and/or orientation of the at least one of the plurality of cameras, with respect to the calibration target repeat steps (i) to (iii), and detecting again the calibration target at the known location in a region of an image of the detection region, and (v) detect, based on an image from at least one of the plurality of cameras, a gaze direction of an eye of the user of the vehicle.
This invention relates to an apparatus for detecting the gaze direction of a user's eye in a vehicle. The system addresses the challenge of accurately tracking gaze direction despite potential misalignment or movement of the cameras used for detection. The apparatus includes an illuminator that directs infrared light toward a detection region encompassing the user's eye and calibration targets positioned at known locations within the vehicle. Multiple cameras capture images of this region, with their fields of view overlapping in a shared detection area. A pattern recognition module identifies the calibration targets within these images. A sensor monitors changes in the position or orientation of the cameras. A control system processes the camera images to determine the pixel coordinates of the calibration targets in two orthogonal directions for each camera. By comparing these coordinates, the system calculates the position and orientation of the cameras relative to the calibration targets. If the sensor detects movement, the system recalibrates by repeating the coordinate determination process. The apparatus then uses the calibrated camera data to detect the user's gaze direction based on the eye's position in the captured images. This ensures accurate gaze tracking even if the cameras shift during vehicle operation.
17. Apparatus for detecting the gaze direction of an eye of a user in a vehicle, the apparatus comprising: an illuminator arranged to direct infrared light towards a detection region, the detection region including the eye of the user and a calibration target at a known location in the vehicle; a plurality of cameras responsive to the light in the detection region and configured to generate an image of the detection region, the plurality of cameras having at least partially overlapping fields of view in an overlap region of the detection region; a controller responsive to the cameras and configured to detect a calibration target at the known location in the overlap region of the image and to determine the position and/or orientation of the at least two cameras relative to the calibration target, wherein the controller detects the calibration target from at least two of the plurality of cameras simultaneously; a sensor configured to detect a change in the position and/or orientation of at least one of the plurality of cameras with respect to the calibration target, and upon a change the controller configured to detect again the calibration target at the known location in a region of an image of the detection region corresponding to an overlapping detection region of at least two of the plurality of cameras, and to determine the position and/or orientation of the at least two cameras relative to the calibration target, wherein the controller detects again the calibration target from at least two of the plurality of cameras simultaneously; the controller further configured to detect, based on an image from at least one of the plurality of cameras, a gaze direction of an eye of the user of the vehicle.
The apparatus detects the gaze direction of a user's eye in a vehicle using infrared light and multiple cameras. The system includes an illuminator that directs infrared light toward a detection region containing the user's eye and a calibration target at a fixed location in the vehicle. Multiple cameras capture images of this region, with their fields of view overlapping in a shared area. A controller processes these images to locate the calibration target in the overlap region, determining the relative position and orientation of at least two cameras. This calibration ensures accurate gaze tracking. If a sensor detects movement of any camera relative to the calibration target, the controller recalibrates by re-detecting the target in the overlapping fields of view of at least two cameras. The system then uses images from one or more cameras to track the user's gaze direction. The apparatus ensures reliable gaze detection by continuously verifying camera alignment with the calibration target, compensating for any positional shifts. This method improves accuracy in vehicle-based eye-tracking applications, such as driver monitoring or human-machine interface interactions.
18. Apparatus according to claim 1 , wherein the plurality of cameras each use a different frequency of modulation.
This invention relates to an apparatus for capturing images using multiple cameras, each operating at a distinct modulation frequency. The apparatus addresses the challenge of interference and signal overlap when multiple cameras capture images simultaneously in the same environment. By assigning different modulation frequencies to each camera, the system ensures that the signals from each camera can be distinguished and processed independently, reducing crosstalk and improving image quality. The apparatus includes a plurality of cameras, each configured to modulate its output signal at a unique frequency, and a processing unit that demodulates the signals to reconstruct the individual images. This approach enables real-time, high-resolution imaging in applications where multiple cameras must operate in close proximity, such as in surveillance, medical imaging, or autonomous vehicle systems. The use of distinct modulation frequencies allows for precise synchronization and separation of signals, enhancing the accuracy and reliability of the captured data. The invention improves upon existing multi-camera systems by mitigating interference and ensuring clear signal differentiation, making it suitable for environments with high electromagnetic noise or dense camera deployments.
19. Apparatus according to claim 6 , wherein the plurality of cameras each use a different frequency of modulation.
This invention relates to a multi-camera imaging system designed to capture high-resolution images or video in challenging lighting conditions, such as low-light or high-contrast environments. The system addresses the problem of limited dynamic range in traditional single-camera setups, which struggle to simultaneously capture bright and dark regions of a scene without overexposure or underexposure. The apparatus includes multiple cameras, each configured to capture images of the same scene but with different exposure settings or modulation frequencies. By using cameras with distinct modulation frequencies, the system can separate and combine the captured data to reconstruct a high-dynamic-range (HDR) image or video. The different modulation frequencies allow the cameras to operate without interference, ensuring that each camera's signal can be independently processed and later merged to produce a final output with improved brightness and contrast. The system may also include processing circuitry to synchronize the cameras, adjust exposure settings, and combine the captured data. The use of multiple cameras with varying modulation frequencies enhances the system's ability to capture detailed images in environments where traditional single-camera systems would fail. This approach is particularly useful in applications such as surveillance, automotive imaging, and scientific imaging, where dynamic range and image quality are critical.
20. Apparatus according to claim 16 , wherein the plurality of cameras each use a different frequency of modulation.
This invention relates to an apparatus for capturing images using multiple cameras, each operating at a distinct modulation frequency. The system addresses the challenge of interference and signal overlap when multiple cameras capture images simultaneously, particularly in environments where electromagnetic interference or overlapping fields could degrade image quality. Each camera in the apparatus is configured to modulate its imaging signals at a unique frequency, allowing the system to distinguish between signals from different cameras. This frequency modulation enables simultaneous operation without interference, improving data acquisition in applications such as surveillance, medical imaging, or industrial inspection. The apparatus may include additional components, such as signal processing units, to demodulate and reconstruct images from the frequency-modulated signals. By assigning distinct modulation frequencies to each camera, the system ensures clear, interference-free image capture, enhancing accuracy and reliability in multi-camera imaging systems.
Unknown
July 7, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.